Response Time Analyzer

Server assisted Performance

Sends a small, rate-limited series of requests (max 20) to a public URL and reports latency statistics (median, percentiles, standard deviation) and a consistency score, with a simple chart.

Privacy: This tool sends the request through the Mutqan backend relay to reach the target. Only the target URL, headers and body you enter are transmitted, and nothing is stored.

What is performance testing?

Performance testing measures how a system behaves under a defined workload: response time (usually the 95th percentile, not the average), throughput (requests per second) and error rate, while watching CPU, memory and database saturation. Load testing uses the expected traffic, stress testing pushes past it to find the breaking point, soak testing holds it for hours to expose leaks, and spike testing checks recovery from a sudden surge.

Read the guide: What is Performance Testing?

How to use Response Time Analyzer

  1. Enter a public http(s) URL (private and loopback addresses are refused by the relay).
  2. Choose the number of samples (1–20) and the method (GET, HEAD or OPTIONS).
  3. Click "Measure" and read the statistics: the median describes typical latency, p95/p99 the tail.
  4. Check the consistency score and the chart for outliers before comparing runs.
  5. Download the samples as CSV or run again to compare before/after a change.

Response Time Analyzer features

  • Sends 1–20 sequential requests through the Mutqan relay and records every sample
  • Min, max, average, median, p90, p95, p99 and standard deviation
  • Consistency score that highlights jitter and cold-start outliers
  • Bar chart of individual samples plus a latency distribution table
  • Status breakdown so failed or rate-limited requests are never hidden in the average
  • Downloadable CSV of the raw samples for reports and dashboards

Response Time Analyzer example

Ten samples against a health endpoint

Input:

URL: https://api.example.com/health
Samples: 10

Output:

Samples: 10
Min: 112 ms · Average: 138 ms · Median: 131 ms
p90: 171 ms · p95: 181 ms · p99: 188 ms · Max: 190 ms
Std deviation: 23 ms (17% of average)
Consistency: 82/100 — stable
Statuses: 200 ×10
In this run the average (138 ms) sits above the median (131 ms), while p95, p99 and the max mark the slow tail that an average hides.

Example run: 10 samples, latency in ms

Axis from 100 ms to 200 ms
  • 112–181 msMin to p95
  • 181–190 msTail: above p95
  1. 1112 ms Min
  2. 2131 ms Median (p50)Half the samples were this fast or faster: the typical latency.
  3. 3138 ms AverageAbove the median, because the slow samples pull it up.
  4. 4171 ms p90
  5. 5181 ms p95Samples slower than this form the tail; SLOs are usually written as p95 or p99 targets.
  6. 6188 ms p99Interpolated between the two slowest samples.
  7. 7190 ms MaxThe slowest sample.

With 10 samples, p99 is interpolated between the two slowest samples, so it lands just below the max: treat it as an indication, not a guarantee.

Frequently asked questions about Response Time Analyzer

How many requests does the analyzer send?

Between 1 and 20 sequential requests through the Mutqan relay, with a timeout per request and per-IP rate limits. It is designed for quick health checks, not sustained load.

What do p90, p95 and p99 mean?

They are percentiles: p95 = 95% of the samples were at least this fast. Percentiles expose tail latency that averages hide, which is why SLOs are usually written as p95 or p99 targets.

Why are my numbers higher than in my own monitoring?

Timings include the network path from the relay to your origin. Compare relative values (before/after a change, endpoint A vs B) rather than absolute milliseconds.

What is the consistency score?

A 0–100 score derived from the coefficient of variation (std deviation ÷ average) and the p95/median ratio. Low scores mean jitter — look for cold starts, caching misses or noisy upstream calls.

Can I test private or localhost endpoints?

No. The relay refuses private, loopback and link-local addresses. Use it only for public URLs you are allowed to test.

Technical notes

Percentiles use linear interpolation between sorted samples, matching the definition used by k6 and most APM products. With a small number of samples p99 is close to the maximum; treat it as an indication rather than a guarantee.

The consistency score blends the coefficient of variation (standard deviation divided by average) with the p95/median ratio. A perfectly stable endpoint scores 100; a p95 three times the median scores 0 on the tail component.

References